---
title: "INSTACART DASHBOARD"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    vertical_layout: fill
    source: embed
---

```{r setup, include=FALSE}
library(tidyverse)
library(p8105.datasets)
library(plotly)

library(flexdashboard)
```

```{r}
# Loading the data
data("instacart")

# Filtering data (sample 10,000 rows)
set.seed(7923) # using UNI as seed
instacart_filtered = instacart %>%
  sample_n(10000)
```

Column {data-width=650}
-----------------------------------------------------------------------

### Chart A
```{r}
# Filtering for top 15 products bought and creating a bar plot
instacart_filtered %>%
  count(product_name, sort = TRUE) %>%
  head(15) %>%
  mutate(product_name = fct_reorder(product_name, n)) %>% 
  plot_ly(x = ~n, y = ~product_name, type = "bar", marker = list(color = ~n, colorscale = "Viridis")) %>%
  layout(title = "Top 15 most ordered products",
         xaxis = list(title = "No. of orders"),
         yaxis = list(title = "Product"))
```

Column {data-width=350}
-----------------------------------------------------------------------

### Chart B

```{r}
# Creating a line plot for average no. of items bought at each hour of the day
instacart_filtered %>%
  group_by(order_id, order_hour_of_day) %>% # grouping each item by order ID
  summarise(items_in_order = n(), .groups = 'drop') %>%
  group_by(order_hour_of_day) %>%
  summarise(avg_items = mean(items_in_order), .groups = 'drop') %>%  # finding mean items bought every hour
  plot_ly(x = ~order_hour_of_day, 
          y = ~avg_items, 
          type = 'scatter',
          mode = 'lines+markers',
        line = list(color = 'black', width = 3),
        marker = list(size = 10, color = 'black')) %>%
  layout(title = "Average items per order by hour of day",
         xaxis = list(title = "Hour of day (24-hour time)", dtick = 2),
         yaxis = list(title = "Average no. of items"))
```

### Chart C

```{r}
instacart_filtered %>%
  mutate(day_name = case_when(
    order_dow == 0 ~ "Sunday",
    order_dow == 1 ~ "Monday",
    order_dow == 2 ~ "Tuesday",
    order_dow == 3 ~ "Wednesday",
    order_dow == 4 ~ "Thursday",
    order_dow == 5 ~ "Friday",
    order_dow == 6 ~ "Saturday"
  )) %>% # assigning week names to codes
  mutate(day_name = factor(day_name, levels = c("Sunday", "Monday", "Tuesday", 
                                                  "Wednesday", "Thursday", "Friday", 
                                                  "Saturday"))) %>%  # ordering data in order of week 
  plot_ly(x = ~day_name, 
        y = ~order_hour_of_day, 
        type = 'box',
        color = ~day_name,
        colors = "viridis") %>%
  layout(title = "Order distribution by hour for each day of week",
         xaxis = list(title = "Day"),
         yaxis = list(title = "Hour of day (24-hour time)"))
```